meadow creek spawning channel - british...

Meadow Creek Spawning Channel - Scarification Monitoring and Suspended

Sediment Assessment 2003-2007

Prepared for Fish and Wildlife Compensation Program, Nelson, BC

Prepared by Darcie Quamme, MSc., R.P. Bio.

Integrated Ecological Research, Nelson, BC

October 2008

Meadow Creek Scarification Monitoring 1

Integrated Ecological Research October 2008

EXECUTIVE SUMMARY

The spawning gravels of the Meadow Creek Spawning Channel (MCSC) are cleaned and redistributed annually using a D6 Caterpillar to dislodge fine sediments (machine scarification). Integrated Ecological Research prepared this report for the Fish and Wildlife Compensation Program in order to: (1) document the procedures that were used to clean settling ponds and restore spawning gravel at MCSC from 2003-2007; (2) determine the effect of scarification on levels of induced turbidity during this time; (3) estimate the total tonnage of sediment removed from the channel; (4) evaluate the effect of re-suspended sediment on fish and fish habitat within the channel and the possible effects on the biological community in Meadow Creek downstream of the channel; and (5) review previous work on gravel studies.

Machine scarification was identified as a cost efficient and effective method for cleaning the spawning gravel. Possible increases in fines within the streambed downstream of the spawning channel following scarification may have negative effects on kokanee spawning habitat. However, scarification occurs before kokanee enter the spawning channel so there is minimal direct effect on adults or incubating eggs.

The enhanced spawning habitat in the channel clearly provides a net benefit to kokanee populations for the Meadow Creek system as a whole. The increased spawning habitat helps sustain Kootenay Lake kokanee, which in turn create a prey base for bull trout and rainbow trout. In addition, a healthy kokanee population generates sport fisheries for these stocks with important financial gains to the surrounding area.

A number of recommendations are made to optimize scarification operations and sediment management techniques in order to mitigate possible impacts.

KEY WORDS Meadow Creek Spawning Channel, scarification, turbidity, suspended sediments, deposited sediment, kokanee



TABLE OF CONTENTS

EXECUTIVE SUMMARY .......................................................................................................................... 1 TABLE OF CONTENTS............................................................................................................................. 2 ACKNOWLEDGEMENTS ......................................................................................................................... 4 1 INTRODUCTION............................................................................................................................... 5

1.1 BACKGROUND.............................................................................................................................. 5 1.2 MONITORING OBJECTIVES............................................................................................................ 6

2 METHODS.......................................................................................................................................... 8 2.1 SCARIFICATION PROCEDURES ...................................................................................................... 8 2.2 TURBIDITY MONITORING ............................................................................................................. 9

2.2.1 Portable field meter ................................................................................................................ 9 2.2.2 Continuous recorder............................................................................................................. 10

2.3 SEDIMENT LOAD IN SPAWNING CHANNEL EFFLUENT ................................................................ 10 3 RESULTS .......................................................................................................................................... 12

3.1 TURBIDITY ................................................................................................................................. 12 3.1.1 Portable field metering of turbidity downstream of spawning channel ................................ 12 3.1.2 Continuous turbidity measurements of spawning channel effluent 2007.............................. 13

3.2 SEDIMENT LOAD ........................................................................................................................ 17 3.2.1 Hourly sediment load based on portable turbidity monitoring............................................. 17 3.2.2 Daily and total sediment load based on continuous meter ................................................... 18

3.3 PREVIOUS GRAVEL STUDIES ..................................................................................................... 20 5 DISCUSSION.................................................................................................................................... 22 6 CONCLUSIONS AND RECOMMENDATIONS .......................................................................... 24 7 LITERATURE CITED .................................................................................................................... 25 LIST OF TABLES Table 1. Timing of final scarification operations at the Meadow Creek Spawning Channel....................... 9 Table 2. Water Survey of Canada (WSC) sites used to estimate discharge. .............................................. 12 Table 3. Summary of continuous turbidity measurements in 2007............................................................ 16 Table 4. Percent mean difference between portable field meter and laboratory measurements ................ 28 Table 5. Percent mean difference between paired laboratory turbidity measurements.............................. 29 Table 6. Percent mean difference between paired laboratory TSS measurements..................................... 30 Table 7. Percent fines and geometric mean diameter of gravels before and after scarification 2005. ...... 32 Table 8. Percent angularity of spawning gravel in MCSC in 2005............................................................ 32



LIST OF FIGURES Figure 1. Location of the Meadow Creek Spawning Channel and overview of the spawning channel...... 7 Figure 2. Turbidity measurements taken with portable field meter……………………………….. ........ 14 Figure 3. Continuous and portable field metered turbidity measurements in 2007. ................................. 15 Figure 4. Continuous turbidity measurements recorded on August 8, 2007............................................. 17 Figure 5. Estimates of hourly sediment load from 2003-2007 ................................................................. 19 Figure 6. Estimates of daily sediment load in 2007 throughout the scarification procedure.. .................. 20 Figure 7. Turbidity measured using portable field meter versus laboratory measurements of TSS .. ...... 27 Figure 8. Turbidity measured using portable field meter versus continuous meter .................................. 31 Figure 9. The cumulative particle size distributions for spawning gravels before and after scarification 33 Figure 10. The number of replicates required to detect a difference in % fines ......................................... 34 LIST OF APPENDICES Appendix 1. Turbidity versus Total suspended sediment (TSS)................................................................ 27 Appendix 2. Quality assurance and quality control of portable field meter. ............................................. 28 Appendix 3. Quality assurance and quality control of laboratory replicates. ............................................ 29 Appendix 4. Turbidity measured by field meter versus paired measurements from continuous meter ..... 31 Appendix 5. Gravel studies data from Porto (2006). ................................................................................. 32 Appendix 6. Cumulative size distributions of spawning gravels. .............................................................. 33 Appendix 7. Minimum sample number requirements per given effect size. ............................................. 34



ACKNOWLEDGEMENTS

James Baxter and Steve Arndt of the Fish and Wildlife Compensation Program carried out the 2007 project initiation and placement of the continuous turbidity meter. James and Steve both provided comments on the draft report.

Murray Pearson collected the data, and supervised the Meadow Creek Spawning Channel operations and scarification work. Murray also reviewed the report.

Jeff Burrows from the Ministry of Environment, Kootenay Fish and Wildlife, collected suspended sediment data and summarized the metering, laboratory data in 2003 and made recommendations on scarification procedures in a previous memo (Burrows 2003).

Darcie Quamme of Integrated Ecological Research carried out data analysis and reporting. John Boulanger, Integrated Ecological Research, gave statistical advice.

The FWCP is a joint initiative between BC Hydro, the Ministry of Environment and Fisheries & Oceans Canada to conserve and enhance fish and wildlife populations affected by the construction of BC Hydro dams in Canada's portion of the Columbia Basin.



1 INTRODUCTION

Integrated Ecological Research prepared this report for the Fish and Wildlife Compensation Program (FWCP) to document the effect of gravel scarification at the Meadow Creek Spawning Channel (MCSC) on the levels of induced fine sediment and its dispersal in the Meadow Creek system downstream of the spawning channel.

1.1 Background

The Meadow Creek spawning channel (Figure 1) was constructed in 1967 in order to partially replace kokanee production and spawning habitat lost when the Duncan River was impounded during the construction of the Duncan Dam (Acara 1970). Kokanee, bull trout, and rainbow trout spawning habitat was impacted by the creation of the dam. Annual fish losses from the construction of the dam were estimated to be 2.8 million kokanee from the Duncan River (Bull 1965). Before flooding Meadow Creek supported 346,128 spawners in 1964 and 116,095 spawners in 1965 (Acara 1970). This was 7.6% (1964) and 10.8% (1965) of the total Lardeau-Duncan River system escapement. After the Duncan River was blocked in 1966, the escapement to Meadow Creek then comprised a larger proportion (46.0% in both 1967 and 1968) of the total escapement to the Lardeau-Duncan River system (Bull 1965; Acara 1970).

The objective of the MCSC is to increase the abundance of kokanee in Kootenay Lake by providing high quality spawning habitat and increased kokanee production through improved egg to fry survival. The operation and maintenance of the spawning channel is funded by a joint initiative of the Ministry of Environment and BC Hydro under the auspices of the FWCP.

The MCSC is located 45 km north of Kaslo at the north end of Kootenay Lake (Figure 1). The spawning channel is supplied with water from diversions on Meadow Creek and John Creek. The spawning channel is 2.9 km long and on average 9.1 m wide with 26,390 m2 of spawning habitat. The theoretical capacity of the channel is 350,000 spawning kokanee with an annual deposition of 45-50 million eggs. It produces between 15-20 million fry annually, with a target egg-to-fry survival rate of greater than thirty-five percent. Throughout the year, naturally occurring fine suspended sediment accumulates in the MCSC due to the low gradient (approximately 0.02%) and reduced water velocities of the channel compared to the natural stream. Also flow control structures within the channel reduce discharge during peak flows. These aspects of the channel design result in the deposition of suspended sediment in the interstitial spaces of the spawning gravel (Mundie and Crabtree 1997). Sediment sources occur, in particular, in John Creek, and directly contribute to the spawning channel (Taylor et al. 1972). An upstream settling pond (approximately



113 m by 13-30 m) was constructed in 1987 to remove some sediment from the water column before it enters the channel.

In order to maintain the quality of spawning gravel and high kokanee egg-to-fry survival, it is important to scarify the gravels annually. This is carried out during a five to six day period in late summer using a Caterpillar D-6 with a push blade that dislodges and re-suspends the fines so that the sediment is flushed downstream of the channel.

The Department of Fisheries and Oceans (DFO) requested that the levels and effects of suspended sediment from the scarification process be summarized. In addition, continuous monitoring of turbidity in 2007 was carried out in order to document possible adverse impacts to fish and recommend mitigation options. This report also provides a record of spot checks of turbidity from 2003-07 during the peak levels of turbidity.

1.2 Monitoring Objectives

The objectives of the review were to:

• Document recent procedures that are used to clean settling ponds and restore spawning gravel at the Meadow Creek Spawning Channel.

• Monitor the effect of scarification on levels of turbidity. • Estimate the total amount of fine material removed from the spawning

channel. • Evaluate the effects of the re-suspended sediments on fish habitat and fish in

the spawning channel and Meadow Creek downstream of the channel. • Review previous work on gravel studies within the spawning channel before

and immediately after cleaning of spawning gravels.



Figure 1. Location of the Meadow Creek Spawning Channel and overview of the spawning channel.



2 METHODS

2.1 Scarification Procedures

In the first week of August (2004-2007) during the instream work window, Stan Baker Trucking carried out the MCSC scarification procedures under the direction of Murray Pearson, the channel manager (Table 1).

Maintenance operations were initiated each year with the removal of suspended bed load that settles in Meadow Creek in the area near the hatchery. Following this, the flow control gate that diverts water to the spawning channel from Meadow Creek was shut down. The gravel berm that develops in front of the settling basin was then removed and the area levelled.

Then in order to dewater and clean the large settling pool upstream of the spawning channel, the flow from the John Creek diversion was redirected to the spawning channel below the settling pond. Typically, 200-700 m3 of sediment are removed from the pond over a period of 3-5 days.

Afterwards, maintenance took place upstream of the spawning channel in John Creek (approximately 100 m upstream of the John Creek diversion fence) where debris accumulated in front of the water diversion. Typically, 60-100 m3 of material were removed from this area.

Finally, scarification of the spawning channel commenced once suspended sediment levels in both John and Meadow creeks returned to baseline levels. In addition, flow through the spawning channel was increased in order to help flush sediment out of the channel. Scarification of the channel typically took five to six days to complete.

Overhead vegetation was removed along each portion of the channel so as not to interfere with scarification. In the channel, the small bulldozer (CAT D6) was used to turn and recast the spawning channel gravel while working from the top of the channel to the bottom each year. The bulldozer then evenly distributed the cleaned gravel.

The bulldozer typically took short breaks of 5-25 minutes every one to two hours so as to reduce turbidity levels for short periods over the scarification workday (Table 1). A 30-40 minute lunch break was also usually taken, as well as the 12-hour period overnight, during which turbidity levels were reduced daily.

From 2003-2007, spot check turbidity monitoring was done during scarification just below the channel outlet (confluence site) and 4 km downstream (highway bridge site) in order to capture the increased turbidity levels resulting from the procedure (Figure 1). In 2007 a continuous monitor was also put in place at the



confluence site to monitor turbidity levels throughout the whole scarification process.

Table 1. Timing of final scarification operations at the Meadow Creek Spawning Channel by

year during peak turbidity levels during scarification of the lower channel.

Year Date Timing of scarification Break (min) Start End 2004 Aug. 7 18:18 19:03 None2004 Aug. 8 7:42 10:32 5 10:37 12:14 5 12:19 18:13 end2004 Aug. 10 6:20 8:10 10 8:20 8:45 23 9:08 11:10 40 11:50 13:50 10 14:00 14:40 15 14:55 15:15 25 15:40 17:05 end2005 Aug. 2 10:30 12:30 20 12:50 1:15 15 1:30 17:00 10 17:10 17:30 end2006 Aug. 6 7:30 9:45 15 10:00 10:15 15 11:00 12:30 30 13:00 13:30 10 13:40 14:00 5 14:05 15:00 15 15:15 16:10 10 16:20 17:00 end2007 Aug. 8 7:00 8:50 10 9:10 11:30 40 12:10 13:40 10 13:50 15:05 5 15:15 17:00 end2007 Aug. 9 7:00 8:50 10 9:00 10:10 end

2.2 Turbidity Monitoring

2.2.1 Portable field meter

From 2003-2007, the spawning channel operator took spot check turbidity measurements each hour with a Lamotte Model 2020 portable turbidity meter during periods of peak turbidity (resolution ±2% 0-100 Nephelometric Turbidity Units (NTU), ±3% above 100 NTU, detection limit 0.5 NTU). Quality



assurance/Quality control (QA/QC) of the turbidity meter was carried out in 2003 (Appendices 1-4).

Monitoring took place just below the confluence of the channel outlet and Meadow Creek (confluence site) and 4 km downstream (highway bridge site) in order to capture the increased turbidity levels resulting from the procedure (Figure 1; Table 2). The portable field meter was calibrated with solutions of 10 NTU before daily measurements at each new location.

2.2.2 Continuous recorder

In 2007, turbidity levels during the scarification were monitored using an Analite NEP495 microprocessor turbidity probe (McVan Instruments Ltd., Australia). The meter was installed downstream of the spawning channel below the confluence of the spawning channel with Meadow Creek (Figure 1, Table 2) on July 13th and turbidity values were recorded until August 18th. The Analite NEP495 has a resolution of ±0.01 at 40 NTU, ±0.02 at 100 NTU, ±0.1 at 400 NTU, and ±0.2 at 1000 NTU for turbidity and is designed to operate in conditions where there are high levels of suspended sediment. The recorder was calibrated prior to installation with solutions provided by the distributor (Geo Scientific Ltd., Vancouver, BC).

The meter was set to log a turbidity measurement once every 20 minutes, and wipe the sampling window prior to each measurement. Additionally, the Analite NEP495 logged a temperature measurement every 20 minutes (resolution ±0.1 from -10 to 50oC).

Data from the Analite NEP495 continuous turbidity meter was higher in thirty-one of the thirty-four paired samples measured using the Lamotte 2020 portable turbidity meter (Appendix 4). As a result, continuous turbidity data was calibrated to the portable field meter for comparison to previous annual monitoring.

2.3 Sediment Load in Spawning Channel Effluent

Hourly sediment loads for 2003-2007 were estimated from spot checks of turbidity using the portable field meter (Lamotte 2020) measured during periods of peak turbidity. Daily and total loads for 2007 were estimated from the automated turbidity metered every 20 minutes using the NEP495. The continuous data turbidity was also calibrated to the Lamotte 2020 meter using the equation found in Appendix 4 prior to load calculations. This was done so that turbidity



values measured in 2007 could be compared to data from past years collected with the portable meter.

Sediment loads were calculated by converting turbidity values in NTU to total suspended sediment (TSS, mg/L) using linear regression of the data from Burrows (2003). See Appendices 1-4 for raw data and quality control.

An inspection of the linear regression of turbidity versus TSS indicated that the intercept value was not significantly different from zero (p=0.64, n=14). When the intercept was included it was negative and thus not possible. In addition, there was minimal data to estimate an intercept because at least twenty data points are required to support an intercept plus slope-term model based on recommendations by Tabachnick, and Fidell. (1996). As a result, the intercept was dropped from the model and the regression was forced through zero on the assumption that when turbidity is below detection (<0.1 NTU), TSS is also near detection (4 mg/L). The confidence intervals for this model widely encompassed the zero value and these detection limits (Appendix 1).

A quadratic term was also tested but was not significant (p=0.60). There was not enough data to support a model with a non-linear term. In addition, laboratory verification of the portable meter at high turbidities was limited to two data points (Appendix 1). Paired TSS-turbidity data across the full range of turbidities is required to test for non-linearity. Previous work by Lewis (1996) also indicates that regressions of suspended sediment concentration versus turbidity are often linear with low variance. As a result, TSS was estimated from the equation below:

TSS = 6.44 x (Turbidity) (r2 = 0.95, p<0.0001) Suspended sediment load was calculated using the equation:

Suspended load = TSS (mg/L) x Discharge (m3/s) x 1,000 L/m3. Load was computed on an hourly, daily or total period basis and milligrams were converted to metric tons.

However, daily discharge data was lacking for 2007 at the monitoring sites and as a result possible ranges in sediment load were estimated based on Meadow Creek flow only and Meadow Creek plus John Creek. Discharge from Meadow Creek was used as a minimum estimate of flow at this site. Also, an unknown percent of flow from John Creek was diverted through the spawning channel. As a result, the discharge of Meadow Creek plus John Creek was used as a maximum range.

Discharge at the confluence of the spawning channel outlet and Meadow Creek (confluence site) was estimated using continuous discharge data from Duhamel and Lemon creeks (Water Survey of Canada archived data). Drainage area ratios (Duhamel:Meadow and Lemon:Meadow creeks) were used to estimate discharge



at the confluence site. The drainage areas and August discharges for Duhamel and Lemon creeks were obtained from Water Survey of Canada records (Table 2).

The drainage areas for the confluence and highway bridge sites were obtained from Water Survey of Canada records (1968-1973 only) by adding drainage areas for Meadow Creek (upstream of John Creek) and for John Creek (just upstream of Meadow Creek) to obtain an estimate of 97.4 km2 (Table 2). The WSC site, Meadow Creek upstream of John Creek, is located 1.61 km upstream from the confluence of the spawning channel outlet and Meadow Creek. The WSC site, John Creek just upstream of Meadow Creek, is approximately about 680 m northwest of the confluence of the spawning channel outlet and Meadow Creek or 300 m upstream of the location where water from the John Creek Diversion can be redirected through the spawning channel.

Using the 97.4 km2 drainage area gave a slight underestimate of discharge and load for the confluence of the spawning channel outlet and Meadow Creek because the site Meadow Creek above John Creek was 1.6 km upstream from this site.

Table 2. Water Survey of Canada (WSC) sites used to estimate discharge and the location of the

suspended sediment monitoring sites. Site WSC Station

ID /Study Site

Latitude Longitude Gross Drainage

Area (km2)

Period Of Record

Obtained

Meadow Ck above John Ck 08NH124 50°16’2’’N 116°59'’1’’E 62.71 1968-1973 John Ck at Meadow Ck 08NH125 50°15’32’’N 116°59'’7’’E 34.72 1968-1973 Duhamel Ck 08NJ026 49°35’44’’N 117°14’42’’E 52.9 2003-2007 Lemon Ck 08NJ160 49o41’52’’N 117 o27’00’’E 178.0 2003-2007 Confluence of MCSC outlet & Meadow Ck confluence site 50°15’14’.83’N 116°59’37.20’’E 97.43 NA Highway Bridge -below confluence of John Ck Diversion & Meadow Ck

highway bridge site

50°15’14’.83’N 116°59’37.20’’E ------ NA

1 689 m upstream of confluence of Meadow Creek and spawning channel outflow, 21.6 km upstream of confluence of spawning channel outlet and Meadow Creek,3 Estimated from adding gross drainage areas of Meadow Creek above John Creek and John Creek at Meadow Creek. 3 RESULTS

3.1 Turbidity

3.1.1 Portable field metering of turbidity downstream of spawning channel

During the years of monitoring (2003-2007) the confluence site (below the confluence of spawning channel water and Meadow Creek), the peak turbidity levels observed in 2004 (2100 NTU) and 2007 (1069 NTU) were the highest of



the period (Figure 2). Lower values were observed in 2006 (850 NTU), and 2005 (257 NTU).

As expected further downstream, peak turbidity levels at the highway bridge site (1.5 km downstream of spawning channel) were generally lower than upstream levels at the confluence site (Figure 2). But again, peak levels were highest in 2004 (632 NTU) and 2007 (308 NTU), with lower levels in 2005 (244 NTU), and 2006 (160 NTU). The maximum turbidity level observed on each day often occurred during the afternoon scarification (Figure 2).

In 2003, monitoring was carried out post-scarification each evening (16:08-20:05) after turbidity returned to baseline to demonstrate that turbidity levels had decreased. As a result the maximum turbidity level in 2003 was much lower (38.7 NTU) at the confluence site and at the highway bridge (38.4 NTU) than in other years

3.1.2 Continuous turbidity measurements of spawning channel effluent 2007

In 2007, the maximum daily turbidity levels during scarification increased from 31.2 NTU to a peak of 637.9 NTU over the course of the nine-day scarification period as the machine worked from the top of the channel downstream (Figure 3, Table 3).

Median daily turbidity levels during scarification ranged from 2.8–40.9 NTU with an overall median of 5.0 NTU (Table 3). The interquartile range for the daily values during scarification varied from 1.6–110.5 NTU with an overall value of 14.7 NTU (Table 3). As the bulldozer worked on the upper legs of the spawning channel fine sediments were flushed downstream with the current. Fines then accumulated in the middle and lower spawning channel.

Efforts were made to decrease the duration of high turbidity levels throughout the workday in order to reduce potential impacts to aquatic life. An example of this is given in Figure 4 for August 8, 2007 in which the machine operators took the following breaks; 8:50-9:10, 11:30-12:10, 13:40-13:50 and 15:05-15:15. As well the 12- to 14-hour period overnight period reduced turbidity to near base levels.



0

300

600

900

8/7 8/8 8/9 8/10 8/11 8/12 8/13

NT

U

2003

0

300

600

900

7:00 10:00 13:00 16:00 19:00 22:00

NT

U August 7, 2006

0

300

600

900

8/7 8/8 8/9 8/10 8/11 8/12

NTU

2004

2100 NTU

0

300

600

900

8/3 8/4 8/5 8/6 8/7 8/8 8/9 8/10

NTU

2007

1069 NTU

1028 NTU

0

300

600

900

7:00 10:00 13:00 16:00 19:00 22:00

NT

U

August 2, 2005

Confluence Hwy Bridge Figure 2. Turbidity measurements taken at two sites with portable field meter on final days of scarification at the MCSC. Blue lines are for the confluence site and red lines are for the highway bridge site.



0

200400

600800

1000

2-Aug3-Aug4-Aug5-Aug6-Aug7-Aug8-Aug9-Aug10-Aug11-Aug12-Aug

NTU

Figure 3. Continuous (blue circles/grey line) and portable field metered (white diamonds/blue line) turbidity measurements in 2007 recorded at the

confluence site, downstream of spawning channel outlet and confluence with Meadow Creek.



Table 3. Summary of continuous turbidity measurements (NTU) in 2007 measured at confluence site during scarification.

Date Median Min Max 90th

percentile IQR1

07/31/07 3.0 0.1 4.1 3.7 0.7 08/01/07 3.0 1.9 5.9 3.9 0.9 08/02/07 3.6 1.9 31.2 9.8 2.7 08/03/07 3.1 1.7 63.0 18.0 1.9 08/04/07 6.6 1.4 215.5 75.3 24.4 08/05/07 2.8 1.9 190.4 5.7 1.6 08/06/07 6.4 1.5 198.5 124.4 28.0 8/07/07 13.9 2.8 254.1 156.3 67.1 08/08/07 40.9 3.2 637.9 271.4 110.5 08/09/07 11.9 2.2 633.8 33.3 13.7 08/10/07 3.5 0.8 256.0 16.6 11.0 08/11/07 1.2 0.5 3.1 1.9 0.6 08/12/07 0.9 0.3 2.4 1.7 0.5 08/13/07 1.0 0.5 4.4 1.7 0.6 08/14/07 0.7 0.1 8.6 1.5 0.6 08/15/07 1.0 0.2 4.2 2.2 1.0 During Scarification 5.0 0.8 637.9 76.4 14.7

Total period 3.0 0.1 637.9 34.4 5.6 1 IQR = Interquartile range, Shaded dates indicate period of scarification



0

200

400

600

4:007:00

10:0013:00

16:0019:00

NT

U

Figure 4. Continuous turbidity measurements recorded on August 8, 2007 (during scarification operations)

at confluence site just downstream of spawning channel and confluence with Meadow Creek. 3.2 Sediment Load

3.2.1 Hourly sediment load based on portable turbidity monitoring

Hourly loads of suspended sediment were estimated from turbidity levels measured with the portable field meter at the confluence site (downstream of the spawning channel outflow and Meadow Creek) (Figure 5). Hourly loads (including background levels) were calculated in order to examine annual differences in load but were not used to estimate daily loads because of limited monitoring. Continuous discharge data from Duhamel and Lemon creeks were prorated by drainage area ratios to obtain daily discharge for this site used in loading calculations.

The highest hourly loads were observed in 2004 downstream of the confluence of John Creek and Meadow Creek (46.3 and 41.4 metric tons estimate based on discharge prorated from Duhamel and Lemon creeks, respectively). Peak hourly loads in 2007 (23.1 and 17.6 metric tons), 2006 (17.9 and 16.4 metric tons), and 2005 (8.7 and 8.2 metric tons) were lower than 2004.

In 2003, monitoring was carried out in the later part of the day (16:32-20:05) to demonstrate that suspended sediment levels dropped after scarification was completed each day. As a result, peak levels in 2003 were much lower (0.19 and 0.21 metric tons based on Duhamel and Lemon creeks, respectively) than in other years.



3.2.2 Daily and total sediment load based on continuous meter

Possible ranges in daily and total sediment load were estimated based on Meadow Creek flow only (minimum) and Meadow Creek plus John Creek (maximum) (Figure 6). Estimates of load were based on the discharge of Meadow Creek (minimum) and discharge of Meadow Creek plus John Creek (maximum) because an unknown amount of flow from John Creek was diverted to the turbidity monitoring location. Load was also estimated using discharge (equation, Section 2.3) from Lemon Creek to prorate the discharge for Meadow Creek from a drainage-area ratio.

The estimated peak daily sediment load occurred on August 8, 2007 was 21.8-134.2 metric tons (including background levels). The peak sediment load removed from the spawning channel effluent (above background) was 21.4 - 33.6 metric tons on August 8.

The estimated total sediment load from August 2-10, 2007 was 67.7-106.3 metric tons (including background levels). The total sediment load removed from the spawning channel effluent (minus background) was 64.2-100.9 metric tons over the entire scarification period. The background suspended sediment load was based on monitoring carried out pre-scarification (August 1) and after levels returned to baseline (August 11-14).

The total load represents the amount of sediment, naturally deposited in the channel, that was re-mobilized and flushed downstream of the channel during scarification. It does not include the amount of sediment directly removed by the excavator from the settling pond. Daily sediment load increased over the scarification period because gravel cleaning began at the upstream end of the channel and slowly worked downstream. Some of the sediment from the upper channel settled at the mid to bottom portions of the channel and required further scarification of the gravels.



0

10

20

30

40

50

8/7 8/8 8/9 8/10 8/11 8/12

m-t

on/h 2003

0

10

20

30

40

50

7:00 10:00 13:00 16:00 19:00 22:00

m-t

on/h

August 7, 2006

0

10

20

30

40

50

8/7 8/8 8/9 8/10 8/11

m-t

on/h

2004

0

10

20

30

40

50

8/3 8/4 8/5 8/6 8/7 8/8 8/9

m-t

on/h

2007

0

10

20

30

40

50

9:00 12:00 15:00 18:00 21:00

m-t

on/h August 2, 2005

Figure 5. Estimates of hourly sediment load (metric tons) at the confluence site based on the portable meter (including background levels). Brown squares indicate that load was estimated using discharge (equation, Section 2.3) from Duhamel Creek to prorate the discharge for Meadow Creek from a drainage-area ratio. Yellow diamonds indicate discharge from Lemon Creek was used.



0

10

20

30

1-Aug3-Aug

5-Aug7-Aug

9-Aug11-Aug

13-Aug

m-to

ns/d

ay

min. max.

Start of Scarification

Last day of scarification

Figure 6. Estimates of daily sediment load (metric tons) removed from the Meadow Creek Spawning

Channel in 2007 (above background). Yellow squares indicate estimates of load (min.) that use the discharge of Meadow Creek to calculate load. Brown squares indicate (max.) estimates of load based on the discharge of Meadow Creek plus John Creek.

3.3 Previous Gravel Studies

The overall spawning habitat quality data described in Porto (2006) for Meadow Creek spawning channel (data in Appendix 5) met BC water quality guidelines for streambed substrate composition (MoE 2008) and recommendations for benthic sediment found in the literature (Chapman 1988 and Kondolf 2000).

The average percent fines <2 mm from samples collected with a hollow core sampler (McNeil and Ahnell 1964) were lower than BC water quality guidelines of 10% at salmonid spawning sites both before (9.1%) and after (7.3%) scarification. In addition, percent fines <2 mm were less than levels of 12-14 % (<1 mm fines) recommended by Chapman (1988) and Kondolf (2000) to maintain good condition of gravel bed material for egg incubation. Average percent fines <6.4 mm found in the Meadow Creek spawning channel met BC water quality guidelines of not more than 25% before (20.7%) and after (19.8%) scarification.

In addition, Kondolf (2000) has shown that the emergence of fry is reduced if fine sediment of 3-6 mm is greater than 30% of bed material. The percent fines of 2-



6.4 mm in the Porto (2006) survey was on average lower than this criterion including values of 11.6% (before scarification) and 12.6% (after scarification).

Kondolf (2000) also recommends that the percentage of fine sediment in spawning gravels be adjusted downward by 33% for percent fines <1 mm and 42% for fines <4 mm to account for the cleaning effect of the spawning fish. The percent fines in the Meadow Creek spawning channel were below recommended standards even before adjustments for the cleaning effect of redd digging. Thus, gravels of the Meadow Creek channel were thought to be of good quality for spawning with regards to fine materials.

The mean geometric diameter of gravel samples met the BC water quality guideline of at least 12 mm at salmonid spawning sites both before (12 mm) and after scarification (13 mm). The mean geometric diameter (dg) is a measure of the central tendency that includes larger framework gravels and relates the whole particle size distribution to salmon embryo survival (Rex and Carmicheal 2002). _______

Dg= √ d16*d84

d16 = 16th percentile, d84 = 84th percentile of particle diameter See Appendix 6 for cumulative size distributions curves before and after scarification.

Porto (2006) found that the percent fines measured from gravel samples collected with a hollow core sampler in the channel were not significantly related to scarification, channel location or an interaction between the two variables (p=0.37). In the present study, a paired sample t-test and power analysis was conducted on this data to further explain the observed results. The null hypothesis that the mean difference between paired before/after samples was zero (no change) was tested for significance. This was done for fines passing <2 mm and 2-6 mm. A nearly significant decrease occurred with percent fines <2 mm at the Meadow Creek spawning channel (p=0.06) and a non-significant increase was observed for percent fines of size 2-6 mm (p = 0.08).

A power analysis (Borenstein et al. 1997 and Zar 1984) of the significant case (percent fines <2 mm) concluded that there was low power (19.7%, n=9, for a two tailed test, α=0.05) to detect a change with the observed effect size of –1.7 and the level of measurement precision (standard deviation of 6.3). It was also determined that >50 replicates using the hollow core sampler would achieve a desired power level of 0.8 for the paired t-test. The hollow core sampler technique may be more appropriate for detecting greater differences in percent fines with lower standard deviation. See Appendix 7 for the relationship between minimum sample requirements and given differences in percent fines.



As well, the difference in percent fines may vary down the length of the channel if scarification is more effective at the top of the channel compared to the mid to lower sections of the channel (Appendix 6). It would require stratified sampling by distance from the top of the channel to account for this effect. Proper stratification of sampling and analysis of variance and covariance statistical models would increase test power through the efficient modeling of differences in percent fines as a function of distance from the top of channel.

Finally, the hollow core sampler is best used at sites where depth and velocities vary by 10-20% among sites and at depths less than 30 cm to prevent loss of fines (Rex and Carmicheal 2002). Site velocity and depth was not discussed in the Porto (2006) study, however, if some sites sampled varied by more that 10-20% or exceeded 30 cm depth it may account for some of the variation observed. Martin et al. (1980) reported an average depth of 30 cm for the Meadow Creek spawning channel and a range of velocities of 0.43-0.73 m/sec.

Porto (2006) also reported that the Meadow Creek spawning channel had more angular and sub-angular gravels in the middle and lower locations sampled within the spawning channel (Appendix 6). The Meadow Creek spawning channel also had lower overall percent (60%) of angular and sub-angular gravels compared to the Hill Creek spawning channel. Compact angular gravel provide poor habitat for spawners, compared to rounder gravels, which are easier for spawners to move and have larger interstial spaces for better egg and alevin development (Porto 2006).

5 DISCUSSION

5.1 Scarification Effectiveness

The operation target of the Meadow Creek Spawning Channel has an annual target of a kokanee egg-to-fry survival rate of greater than 35%. The spawning channel improves egg-to-fry survival rates from 2-10 % in the natural stream to 20-60% in the spawning channel (Burrows 2003). From 2002-2006, kokanee egg-to-fry survival rate averaged 47% (Spence et al. 2006) in part due to successful scarification techniques, high quality gravels and careful management of the spawning channel operations. The spawning channel is expected to continue to have high egg-to-fry survival rates in the coming years (Spence et al. 2006).

The effective removal of sediments from the spawning channel is essential for achieving kokanee production targets for the facility. Annual machine scarification of gravels and flushing of suspended sediment sufficiently cleans the gravel to allow a high egg-to-fry survival over the following incubation period.



This method is logistically and economically feasible and much less costly than dry screening (Thompson 2006).

The amount of sediment that moved downstream of the channel during scarification was estimated to be 67.7-106.3 metric tons (total including natural background load) and 64.2-100.9 metric tons (above background, remobilized sediment from channel). In addition, typically 200-700 m3 of sediment and organic material were removed from the upstream settling pond. Estimates of total sediment load in effluent at Meadow Creek spawning channel (channel area 26,390 m2) bracketed previous values reported by Mundie and Crabtree (1997) for Little Qualicum (82 metric tons, 31,783 m2) and Quamme and Arndt (2007) for Hill Creek (total load 74 metric tons, 14,888 m2) spawning channels.

5.2 Downstream Effects

The scarification procedure does not add sediment to the creek but changes the timing of movement of sediment through the system. Sediment settles in spawning gravels during most of the year and then is re-mobilized by scarification. The sediments flushed downstream during scarification are part of the annual sediment load carried by Meadow Creek and John Creek. Thus, spawning channels do not add extra sediment load to a stream unlike other impacts such as road building, linear developments, and activities related to agriculture, forestry, mining or urban growth. In addition, any sediments settling just downstream of the spawning channel after scarification in late summer will be mobilized and carried downstream by the following spring freshet.

It is possible that scarification may potentially have some effects on downstream populations of resident species of fish such as non-native brook trout, and native bull trout, rainbow trout, and west slope cutthroat trout in Meadow Creek. However, the effects are thought to be minimal because of low quality rearing habitat in Meadow Creek downstream of the channel (Murray Pearson, pers. com.).

Fine sediments resulting from scarification could also potentially have some detrimental effects on incoming kokanee or the few spawning gravels downstream of the channel. However, these detrimental effects are more than compensated for by enhanced egg-to-fry survival in the spawning channel. Thus, there is clearly a net benefit for the overall kokanee population and their spawning habitat.



6 CONCLUSIONS AND RECOMMENDATIONS The Meadow Creek Spawning Channel clearly compensates for lost kokanee spawning habitat from the construction of Duncan Dam and results in an overall gain in available spawning production. It helps maintain Kootenay Lake kokanee abundance, which provides a forage base for larger predator species including the Gerrard rainbow trout and bull trout. These fish stocks contribute to a popular sport fishery, which has important economic benefits for the local area. Thus, the operation and maintenance of the channel is important for the entire Kootenay Lake ecosystem.

Machine scarification is an effective and affordable technique for annual cleaning of the spawning gravel. In addition, monitoring of fines within the channel (Porto 2006) suggested that gravels of the Meadow Creek channel are of high quality for spawning kokanee. Scarification does not add sediment to the creek but changes the timing of sediment mobilization through the spawning channel and natural stream.

Continued turbidity monitoring should be carried out using both the continuous turbidity recorder and the portable field meter to carry out spot checks upstream and downstream of the Meadow Creek Spawning Channel. The continuous turbidity recorder allows calculation of the amount of sediment removed from the spawning channel and can be used to assess sediment management techniques. Development of a stage-discharge relationship for the spawning channel gauge is recommended during scarification in order to improve sediment load calculations in the effluent.

Quality control procedures for the continuous meter should follow White (1999) and Butcher and Gregory (2006). Quantification of the relationship of TSS versus turbidity is needed across the full range of turbidity on a yearly basis due to possible errors associated with continuous meters (Butcher and Gregory 2006).



7 LITERATURE CITED Acara, A.H. 1970. The Meadow Creek Spawning Channel. Dept. Recreation and

Conservation, Fish and Wildlife, Victoria, B.C.

Borenstein, M.,H. Rothstein, and J. Cohen. 1997. Power and Precision. Version 1. Washington, DC, USA: National Institute of Mental Health.

Bull, C.J. 1964. Estimation of kokanee salmon populations on the Duncan-Lardeau River system. 1964. Fish and Game Branch, Department of Recreation and Conservation, Victoria, B.C.

Burrows, J. 2003. Scarification and total suspended solids at Kokanee, Redfish and Meadow Creek spawning channels. Memo. 10 p. Ministry of Environment, Fisheries Branch, Nelson, B.C.

Butcher, G.A. and L. Gregory. 2006. Continuous water sampling procedures: operating procedures. Prep. By Watershed Aquifer Science, Science and Information Branch. Prep. For Resource Information Standards Committee.

Cavanagh, N., R.N. Nordin, L.G. Swain and L.W. Pommen 1998. Ambient Fresh Water and Effluent Sampling Manual. Water Quality Branch, Environmental Protection, Environment, Victoria, .

Chapman, D.W. 1988. Critical review of variables to define the effects of fines in redds of large salmonids. Transactions of the American Fisheries Society 117:1-21.

Kondolf, G.M. 2000. Assessing salmonid spawning gravel quality. Transaction of the American Fisheries Society 129:262-281.

Lewis, J. 1996. Turbidity –controlled suspended sediment sampling for runoff-event load estimation. Water Resources Research 26 (7), 229-2310.

Martin, A.D., J. Hutchinson, J. Bell and J.L. Fleck. 1987. Scarification, spawning gravel quality, and kokanee production from Meadow Creek Spawning channel 1986-1987. Kootenay Region Report No. 29, Fisheries Branch.

McNeil, W.J. and W.H. Ahnell. 1964. Success of pink salmon spawning relative to size of spawning bed materials. U.S. Fish and Wildlife Service Special Scientific Report Fisheries 469.

MoE. 2008. http://www.env.gov.bc.ca/wat/wq/index.html#criteria

Mundie, J.H. and D.G. Crabtree. 1997. Effects on sediments and biota of cleaning a salmonid spawning channel. Fisheries Management and Ecology 4: 111-126.



Porto. L. 2006. Hill Creek Spawning Channel Scarification and Gravel-Related Monitoring Activities - 2005. Fish & Wildlife Compensation Program, File 145-17. 29 p. + appendices.

Quamme. D. and S. Arndt. 2007. Hill Creek spawning channel – Scarification and suspended sediment assessment. Prep. for Fish & Wildlife Compensation Program. 33 p. + appendices

Rex, J.F. and N.B. Carmicheal. 2002. Guidelines for monitoring fines sediment deposition in streams. Prep. By BC Ministry of Water, Land and Air Protection for the Resource Information and Standards Committee.

Spence, C., V. Ewert and J. Burrows. 2006. Hill and Meadow Creek spawning channel operations and maintenance review and options. Ministry of Environment, Fisheries Branch, , Nelson, B.C.

Tabachnick, B. G., and L. S. Fidell. 1996. Using multivariate statistics. Harper Collins, New York.

Taylor, G.D., D.R. Hjorth, and C. Whitman. 1972. A review of physical, chemical and biological parameters as they relate to kokanee production, particularly in Meadow Creek spawning channel. Fish Habitat Improvement Section, B.C. Fish and Wildlife Branch, Victoria, B.C.

Thompson, A. 2006. Reduction of spawning gravel sedimentation and treatment of scarification effluent at Hill Creek Spawning Channel. Report prepared by Mountain Station Consultants Inc. for Columbia Basin Fish and Wildlife Compensation Program. 45 p. + appendices.

White, E.T. 1999. Automated water quality monitoring. Field Manual. Ministry of Environment, Lands and Parks, Water Management Branch, Water Quality Section.

Zar, J.H. 1984. Biostatistical Analysis 2nd Ed. Prentice-Hall Inc. Englewodd Cliffs, New Jersey. U.S.A.



APPENDICES

Appendix 1. Turbidity versus Total Suspended Solids (TSS).

y = 6.44*x, R2=0.95

TSS

(mg/

L)

-1000

0

1000

2000

3000

4000

5000

Metered turbidity (NTU)0 100 200 300 400 500 600 700

Figure 7. Blue circles indicate the turbidity measured using portable field meter (Lamotte 2020) versus laboratory measurements of TSS from 2003 used to estimate predicted TSS (TSSP). TSSP= 6.44 * (Turbidity) (r2 = 0.95, p<0.001). Black open circles indicate TSSP based on turbidity from the continuous meter (NEP495) for 2007. .



Appendix 2. Quality assurance and quality control of portable field meter.

Table 4. Percent mean difference between paired turbidity measurements taken with portable field meter in 2003 and laboratory measurements

Field meter vs lab turbidity (NTU) Field

(n=1) Lab

( n=2) Mean % dif

0.380 0.455Kokanee 152.000 225.000 18.0 86.500 62.300 38.7 52.000 47.050 32.5 45.000 21.950 10.0 0.290 0.300 68.9Redfish 28.700 17.700 3.4 45.000 38.350 47.4 85.000 52.450 16.0 516.000 527.000 47.4 1.470 4.000 2.1Meadow 33.000 42.450 92.5

36.600 39.250 25.0 65.100 64.300 7.0 2528.000 2150.000 1.2 16.2 Overall mean 28.4



Appendix 3. Quality assurance and quality control of laboratory replicates.

Table 5. Percent mean difference between paired turbidity measurements taken in the laboratory from each water sample collected in 2003

Turbidity NTU Mean % dif. Replicate 1 Replicate 2

Kokanee 0.52 0.39 28.6 190 260 31.1 55.9 68.7 20.5 50.8 43.3 15.9 21.5 22.4 4.1 Redfish 0.33 0.27 20.0 532 522 1.9 44.5 60.4 30.3 35.1 41.6 16.9 16.1 19.3 18.1 Meadow 0.65 0.64 1.6 1900 2400 23.3 69.9 58.7 17.4 52.8 32.1 48.8

39.9 38.6 3.3 Overall mean

18.8

Recommended difference when >five times detection (0.5 NTU)1

25%

1Cavanagh et al. 1998



Table 6. Percent mean difference between paired TSS measurements taken

in the laboratory from each water sample collected in 2003 TSS mg/L Mean % dif.

Replicate 1 Replicate 2 Kokanee 4 4 0.0 1370 1830 28.8 272 368 30.0 264 197 29.1 97 120 21.2 Redfish 9 4 76.9 3020 3470 13.9 420 411 2.2 337 336 0.3 166 180 8.1 Meadow 5 4 22.2 9440 10000 5.8 277 250 10.2 255 172 38.9

176 184 4.4 Overall mean

19.5

Recommended difference when >five times detection (1 mg/L) 1

25%

1Cavanagh et al. 1998



Appendix 4. Paired turbidity measurements from the automated and portable field meters.

y = 0.6983x - 29.712R2 = 0.595

0

300

600

900

1200

0 300 600 900 1200Automated NTU

Fiel

d m

eter

NT

U

1:1 Line

Figure 8. Turbidity measured using portable field meter versus continuous automated turbidity measurements (p<0.001). Automated data points are a mean of two values that bracket the field metered values unless automated turbidity values were metered within five minutes of field metered point then the closest automated value is used.



Appendix 5. Gravel studies data from Porto (2006).

Table 7. Percent fines and geometric mean diameter of spawning gravel

before and after scarification 20051.

Scarification period

Channel location

Sample location across channel

% Fines passing Total % <6.4

Geometric Mean Diameter (dg)

<2mm 2-6.4 mm

Before scarification

Upper Left Center Right

12.7 10.4 11.5

14.7 18.3 16.6

27.4 28.7 28.1

8.66 9.00 7.75

Middle Left Center Right

7.1 10.8 12.4

7.6 13.6 12

14.7 24.4 24.4

15.43 10.20 8.31

Lower Left Center Right

5.7 7.7 3.3

8.4 6.9 6.6

14.1 14.6 9.9

14.49 14.97 14.58

Mean 9.1 11.6 20.7 11.5 After scarification

Upper Left Center Right

4.3 4.2 5.1

10.6 11.5 17.3

14.9 15.7 22.4

13.23 13.86 11.18

Middle Left Center Right

6.0 21.8 8.6

9.7 22.1 14.7

15.7 43.9 23.3

13.86 4.86 11.62

Lower Left Center Right

7.9 0.9 6.8

12.9 4.5 9.7

20.8 5.4 16.5

11.62 20.00 13.86

Mean 7.3 12.6 19.8 12.7 1Table taken from Porto (2006).

Table 8. Percent angularity of spawning gravel in MCSC in 20051.

Location % Angularity % Angularity of Total

Angular Sub-Angular

Round Sub-round

Upper 11 49 10 30 60 Middle 13 51 13 23 64 Lower 11 41 15 33 52 Mean 59

1Table taken from Porto (2006).



Appendix 6. Cumulative size distributions of spawning gravels.

0

20

40

60

80

100%

Pas

sing

Upper Channel

50th percentile

BeforeAfter16th percentile

84th percentile

0

20

40

60

80

100

% P

assi

ng

Mid Channel

50th percentile

BeforeAfter

84th percentile

16th percentile

0

20

40

60

80

100

0.01 0.1 1 10 100Grain size (mm)

% P

assin

g

Lower Channel

50th percentile

BeforeAfter

16th percentile

84th percentile

Figure 9. The cumulative particle size distributions for spawning gravels before and after scarification at

upper, mid and lower locations within the channel (n=3 for each location and time period). The 84th and 16th percentiles are used to calculate the mean geometric diameter (see Section 3.3)



Appendix 7. Minimum sample number requirements to detect a difference in % fines with a paired t-test.

0102030405060

2 3 4 5 6Effect size (% Fines)

#Rep

licat

es

Figure 10. The number of replicates required to detect a difference in % fines (effect size) for a standard deviation =5.5. The standard deviation used here was an average of the observed standard deviations measured for size fractions of <2mm and 2-6 mm percent fines.

meadow creek spawning channel - british...

Documents